Originally posted 2013-04-12 19:32:08.
Four key factors in photography are linked: aperture, shutter speed, film or sensor sensitivity and subject brightness. These are linked in such a way that if any one is changed, then one or more of the others must also be changed to compensate. The first two, aperture and shutter speed, are easily controlled on camera, and now we will look at the third factor, which is also under the photographer’s control: film or sensor sensitivity.
In pre-digital days this was always referred to as ‘film speed’ and this is convenient; we continue to use it to describe digital sensor sensitivity — sensor speed. A number of indices or scales were established to measure film speed, but the one that became the most widely used was the American Standards Association, or ASA, scale, which was eventually adopted by the International Standards Organisation or ISO. This is an arithmetical scale, so that a doubling of the ISO value indicates a doubling of sensitivity or speed, and a halving the opposite. (This is as opposed to the more complex DIN scale which was logarithmic.) Because the ISO values are arranged in this way, they conform exactly to the conventions we have already met for aperture and shutter speeds, where each stop or step is a doubling or halving of the one before. So it is easy to apply the same logic to film/sensor speed.
Here is an example: working in low light, we find that in order to get the correct exposure, we need to set an aperture of f2 and a shutter speed of 1/15th of a second with ISO 100 film. However, this is too slow a shutter speed for the subject, and we are hand-holding. The lens is already at its maximum aperture. In this case we could swap the ISO 100 film for, say ISO 400. This is a speed increase of 2 stops, which lets us increase the shutter speed (reduce the time it is open) by a reciprocal 2 stops, meaning that we could use a shutter speed of 1/60 of a second. If this would still not be enough then we might further raise the film speed. We can do exactly the same with digital just by adjusting a knob!
However, with both film and digital, there is a trade-off in quality when we raise the film speed and this has to be considered. In film, there are two factors which cause a loss of quality. Essentially both are caused by the fact that in a more sensitive film, more light-sensitive silver salt is required, and the emulsion has to be thicker. The first factor we encounter is film grain or graininess, and the second is that with increased emulsion thickness there is more scope for light to scatter as it passes through, affecting silver molecules which are not part of the image. So in addition to increased graininess, we see a degradation of sharpness.
Digital sensor speed
Digital sensors also have limitations to speed, but these are nowhere near as marked as with film. In fact, one of the big digital advantages that was spotted early on by press and sports photographers was that digital sensors were far better in low light and when underexposed than film. However, that does not mean that everything is perfect and we can increase speed as far as we like, because the more a digital sensor’s sensitivity is pushed, the more digital noise becomes a problem. Essentially this limits the extent to which digital can be pushed, to similar levels as film, with the advantage, however, that the digital sensor will perform much better than film where there is a lot of under-exposure.
Modern ‘full-size’ 35mm equivalent sensors, which are 36mm x 24mm, are really remarkably in this regard and can be pushed to their limits with minimal loss quality, where film would collapse into what we used to call ‘soot and whitewash’.
Film users also increase the effective speed of film by extending the development or ‘push-processing’ film. However, in addition to the problems we have seen, this causes a marked increase in film contrast. This disadvantage does not apply to digital, another reason why it became so quickly accepted by news and sport professionals like me.
In the days of film it was axiomatic that the slower the film the higher the quality, for the reasons seen above. However, most users made a compromise judgement that balanced maximum quality against practicality. This is why film like Ilford’s Pan F and the Kodak equivalent were never as popular as their mid-range products like FP4 or Plus-X. Typically the slower films were in the speed range around 50 ISO, and the mid-range 100-125 ISO. In colour, the fact that there have to be at least three emulsion layers, each sensitive to a different colour of light, red, green or blue, makes the problem worse, hence the professional’s insistence on films like Kodachrome 25 and Fuji Velvia for high quality work. (Fuji actually developed a four-layer emulsion.)
However, the digital sensor, not having to deal with graininess, does not have to be so slow and modern digital sensors will show no appreciable noise up to 200 or 400 ISO, or even more. SO within a considerable range of sensitivity the sensitivity can be chosen to suit the desired effect with no visible degradation. This gives the photographer tremendous flexibility.
The choice of sensor speed can therefore be more dependent on other factors, such as the combination of aperture and shutter speed that we want to use. For example, if we wanted to isolate the subject but also allow some motion blur, perhaps in a picture of someone running, then we might want a wide aperture and a slow shutter speed. The answer is to reduce the sensor sensitivity or film speed. (Once, the use of Neutral Density or ND filters was popular to allow this without changing film, but in a digital age these are redundant.) On the other hand, if we want depth of field and to stop motion, then we can increase sensor speed to allow a smaller aperture and faster shutter speed, though bearing in mind the limitations above. On a digital SLR this function is easy and quick to use.